Browse by author
Lookup NU author(s): Dr Zeng-ping Chen, Professor Elaine Martin
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
The successful resolution of overlapping chromatographic peaks using multivariate resolution methods depends on information about the concentration windows of all the chemical components being available. When conventional methods based on evolving factor analysis are used to obtain such information, a problem arises as a consequence of there not being a one-to-one correspondence between the rank maps and the elution patterns. The recursive evolving spectral projection (RESP) methodology developed in this paper offers a solution to this problem. The algorithm is initialized with a gross partition of the whole retention time region into several exclusive forward or backward informative regions. After selecting a representative spectrum from every informative region, the concentration windows of the overlapping components are determined by examining the residual plots obtained from the proposed recursive evolving spectral projection procedure. Both theoretical and experimental results demonstrate that RESP possesses the characteristics of easy implementation and good performance, which will help realize the automation of non-iterative multivariate resolution methods. © 2004 Elsevier B.V. All rights reserved.
Author(s): Chen ZP, Morris J, Martin E, Yu R-Q, Liang Y-Z, Gong F
Publication type: Article
Publication status: Published
Journal: Chemometrics and Intelligent Laboratory Systems
Year: 2004
Volume: 72
Issue: 1
Pages: 9-19
ISSN (print): 0169-7439
ISSN (electronic): 1873-3239
Publisher: Elsevier BV
URL: http://dx.doi.org/10.1016/j.chemolab.2004.02.001
DOI: 10.1016/j.chemolab.2004.02.001
Altmetrics provided by Altmetric
